This invention relates to amidine, amidrazone and amidoxime derivatives of monosaccharides which have an unexpectedly broad spectrum glycosidase inhibitory effect. The invention further relates to methods of preparing the derivatives, novel intermediates, methods of preparing the intermediates, and the use of the derivatives to inhibit glycosidases, particularly multiple classes thereof.
Glycosidases are enzymes that catalyze the hydrolysis of glycosidic bonds and are essential for the normal growth and development of all organisms. Their vital role is reflected in their wide distribution in nature. They participate in biologically significant reactions such as the breakdown of carbohydrate foodstuffs, the processing of eucaryotic glycoproteins, the catabolism of polysaccharides and glycoconjugates, and the like. Certain glycosidase inhibitors have been found to inhibit human immunodeficiency virus (HIV) syncytium formation and virus replication, thereby indicating their potential use as antiretroviral agents. And some glycosidases are showing promise in treating diabetes or as antiviral and anticancer agents.
Polyhydroxylated piperidines constitute a class of naturally-occurring glycosidase inhibitors. Examples of such monosaccharide analogs which contain a nitrogen atom (an endocyclic nitrogen) in place of the pyranose oxygen include: nojirimycin (A) and 1-deoxynojirimycin (B) (Inouye et al., Tetrahedron, 1968, 24, 21252144), 1-deoxymannojirimycin (C) (Fellows et al, J. Chem. Soc. Chem. Commun., 1979, 977-8), and galactostatin (D) (Miyake et al., Agric. Biol. Chem., 1988, 52, 661-6). These alkaloids are quite specific inhibitors of their targeted enzymes. The arrangement of the hydroxyl groups apparently determines individual enzyme specificity of these inhibitors. Compounds A and B are potent glucosidase inhibitors that closely resemble glucose; compound C inhibits mannosidases; and compound D inhibits galactosidases. The inhibitors are believed to function by mimicking the natural substrates. Compound B has been shown to interfere with the infectivity of HIV (Gruters et al., Nature, 1987, 330, 74-7).
Since a number of glycosidases are insensitive to the naturally occurring alkaloids, tremendous effort has been devoted to the development of synthetic inhibitors which may inhibit those enzymes. Examples include inhibitors of: jack bean .alpha.-mannosidase (E) (Eis et al., Tetrahedron Lett, 1985, 26, 5397-8), coffee bean .alpha.-galactosidase (F) (Bernotas et al., Carbohydr. Res., 1987, 167, 305-11), .alpha.-N-acetylglucoaminidase (G) (Fleet et al., Chem. Lett., 1986, 1051-4), and .alpha.-hexosaminidase (H) (Bernotas et al., Carbohydr. Res., 1987, 167, 312-6). In these polyhydroxylated piperidines, the ring oxygen in the sugar has been replaced by a nitrogen to form azasugars. Unfortunately, the analogy of using azasugars as glycosidase inhibitors is not always applicable. For example, while polyhydroxylated pyrrolidines are better inhibitors of yeast .alpha.-glucosidase, their relative activities are reversed for a number of mouse gut disaccharides. Therefore, the actual effectiveness of a given compound as a glycosidase inhibitor of a specific enzyme and the effect on specificity by a given structural change of the compound remain unpredictable.
Other potent glycosidase inhibitors contain at the anomeric position an exocyclic nitrogen. (Lai et al., Biochem. Biophys. Res. Commun., 1973, 54, 463-8) For example, glucosylamine inhibits both .alpha.- and .beta.-glucosidases.
The present invention is the result of attempting to combine into a single compound several specific features which, among many others, have been found to be present in some glycosidase inhibitors to determine whether the combination compound would prove beneficial. Thus, the compounds of this invention contain both endocyclic and exocyclic nitrogens in an sp.sup.2 -hybridized functional group and also have a flattened, half-chair conformation. While previous glycosidase inhibitors have generally shown activity only against a single class of glycosidases, i.e. glucosidases, mannosidases, galactosidases, etc., they have not demonstrated broad spectrum activity across multiple classes. The amidine, amidrazone, and amidoxime derivatives of the monosaccharide analogs of the present invention, on the other hand, have been unexpectedly found to exhibit an apparently unique and broad spectrum of glycosidase inhibitory activity which extends across class lines, i.e. independent of the sugar analog from which they are formed.
Previous attempts at preparing amidine derivatives of glucose to form possible glycosidase inhibitors were reported by Bird et al., Can. J. Chem., 1990, 68(2), 317-22. Bird et al. started with glucose and prepared 5-azido-2,3,4,6-tetra-O-benzyl-5-deoxy-D-gluconitrile and 2,3,4,6-tetra-O-benzyl-5-deoxy-5-trifluoroacetamido-D-gluconitrile, but was unable to convert these compounds into a protected 5-amino-5-deoxy-D-gluconitrile and to subsequently cyclize them to an amidine analog of glucose.
Accordingly, it is an object of the present invention to produce glycosidase inhibitors which combine both endocyclic and exocyclic nitrogens into an sp.sup.2 -hybridized functional group and also have a half-chair conformation.
It is a further object to produce novel amidine, amidrazone, and amidoxime derivatives of monosaccharides.
It is a further object to develop a process for producing glycosidase inhibiting compounds which process could generate amidines, amidrazones or amidoximes of different glycoses by simply changing the configuration of the starting material.
These and still further objects will be apparent from the following detailed description of the invention.